Clinical Significance of the Arthroscopic Drive-Through

Sign in Shoulder Surgery

Edward G. McFarland, M.D., Carlos A. Neira, M.D., Maria Isabel Gutierrez, M.D., MSc.,
Andrew J. Cosgarea M.D., and Mike Magee, M.D.

Purpose: During arthroscopy of the shoulder, the ability to pass the arthroscope easily between the
humeral head and the glenoid at the level of the anterior band of the inferior glenohumeral ligament
is considered a positive drive-through sign. The drive-through sign has been considered diagnostic of
shoulder instability and has been associated with shoulder laxity and with SLAP lesions. The goal of
this study was to examine the prevalence of the drive-through sign in patients undergoing shoulder
arthroscopy and to determine its relationship to shoulder instability, shoulder laxity, and to SLAP
lesions. Type of Study: Case series. Methods: We prospectively studied 339 patients undergoing
arthroscopy of the shoulder for a variety of diagnosis from 1992 to 1998. The drive-through sign was
performed with the patients in a lateral decubitus position and under general anesthesia. The
drive-through sign was correlated with preoperative physical findings, intraoperative laxity testing,
and with intra-articular pathology at the time of arthroscopy. Results: The arthroscopic evaluation
showed that drive-through sign was positive in 234 (69%) shoulders. For the diagnosis of instability,
the drive-through sign had a sensitivity of 92%, a specificity of 37.6%, a positive predictive value of
29.9%, a negative predictive value of 94.2%, and an overall accuracy of 49%. There was an
association between the drive-through sign and increasing shoulder laxity, but not with SLAP lesions.
Conclusions: This study shows that a positive drive-through sign is not specific for shoulder
instability but is associated with shoulder laxity. This arthroscopic sign should be incorporated with
other factors when considering the diagnosis of instability. Key Words: Instability—Arthroscopy—
Labrum—Laxity—SLAP lesions—Shoulder.

T he diagnosis of instability of the shoulder de-

pends on a thorough history and physical exam-
ination. The vast majority of patients with instability
have a history of dislocation or subluxation of the
shoulder that is either confirmed radiographically or
by provocative maneuvers in the office. Signs of
shoulder laxity on physical examination include a
From the Department of Orthopedic Surgery, Division of Sports
Medicine and Shoulder Surgery (E.G.M., C.A.N., A.J.C.) and the
Department of Mental Hygiene, School of Public Health (M.I.G.),
The Johns Hopkins University, Baltimore, Maryland; and private
practice, Prince George’s Orthopaedic Associates, Clinton, Mary-
land (M.M.), U.S.A.
Address correspondence and reprint requests to Edward G.
McFarland, M.D., 10753 Falls Rd, Suite 215, Lutherville, MD
21093, U.S.A. E-mail: emcfarl@jhmi.edu
© 2001 by the Arthroscopy Association of North America
0749-8063/01/1701-2402$35.00/0
doi:10.1053/jars.2001.19967
positive anterior and posterior drawer, a load and shift
test, and a sulcus sign. Signs of shoulder instability on
physical examination have been well described in the
literature and include the apprehension test, the pos-
terior apprehension test, the Feagin test, and the relo-
cation test.1-7
In some patients, the diagnosis is unclear because of
an ambiguous history or an equivocal physical exam-
ination. This is particularly true in some athletic
individuals who have no specific finding on physi-
cal examination to confirm their diagnosis of insta-
bility.3,8-10 These patients often present with pain,
dead arm symptoms, or both, and imaging modalities
may not offer any assistance in making the diagno-
sis.1,2,4,10 In this group of patients, diagnostic arthro-
scopy may be of assistance in delineating the pathol-
ogy. Intraoperative findings suggestive of instability
include an anterior or posterior labral injury or avul-

38 Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 17, No 1 (January), 2001: pp 38 – 43
 ARTHROSCOPIC DRIVE-THROUGH SIGN
sion (i.e., Bankart lesion), a Hill-Sachs lesion, a cap-
sular rent or tear, glenoid rim erosions, or partial
thickness rotator cuff tear.
Another arthroscopic test postulated to be associ-
ated with shoulder instability is the drive-through
sign. This test is performed by pushing the arthro-
scope through the shoulder between the head of the
humerus and the glenoid. The drive-through sign has
been described by Pagnani and Warren11-13 as the
ability to lever the arthroscope between the glenoid
and the humeral head in the region of the anterior band
of the inferior glenohumeral ligament. In a group of
patients who underwent an anterior capsulolabral re-
construction for painful instability, Jobe et al.14 re-
ported that the only sign found in all of the patients
arthroscopically was a positive drive-through sign.
The drive-through sign has also been postulated to
reflect ligamentous laxity of the shoulder.11-13,15 In
patients undergoing arthroscopic stabilization of the
shoulder, elimination of the drive-through sign has
been considered indicative of a successful tightening
of the glenohumeral ligaments. Morgan et al.15 have
suggested that the drive-through sign is associated
with SLAP lesions, partial rotator cuff tears, and pos-
terior-superior instability. He found that repair of the
SLAP lesions eliminated the drive-through sign.
To our knowledge, the drive-through sign has not
been extensively studied and there are no previous
reports that address its presence in a group of patients
with shoulder problems. The goal of this study was to
examine the prevalence of this sign in patients under-
going shoulder arthroscopy and to determine its rela-
tionship to shoulder instability, shoulder laxity, and
intra-articular shoulder pathology. The relationship of
this sign to other preoperative findings on physical
examination was also considered to be important. It
was hypothesized that the drive-through sign would be
associated highly with shoulder laxity and instability,
but that it would not be diagnostic of shoulder insta-
bility.
METHODS
Between 1992 and 1998, 349 patients who under-
went diagnostic arthroscopy of the shoulder were pro-
spectively entered into this study. There were 10 pa-
tients in whom the drive-through sign was not
documented or not performed, so 339 patients were
included in the analysis. In all patients, a thorough
preoperative physical examination was performed that
included range of motion of the affected and unaf-
39
fected shoulders, instability tests, impingement signs,
and a neurologic examination.
All patients underwent shoulder arthroscopy under
general anesthesia with a scalene block or with intra-
venous neuromuscular blocking agents, and no pa-
tients were operated on with only intravenous seda-
tion. After the induction of anesthesia, both shoulders
were examined for shoulder laxity with the patient in
a supine position. Anterior and posterior laxity was
graded using a modified Hawkins classification as
previously described.16 A sulcus test was performed
on each shoulder and reported using a standard grad-
ing scale (I ⬍1.0 cm, II 1.0 to 1.5 cm, III ⬎1.5
cm).1,2,4,14,16-18 Patients with frozen shoulders had ar-
throscopy and laxity testing after a closed manipula-
tion of the shoulder.
Arthroscopy was performed with the patients in a
lateral decubitus position and the arm held in an arm
holder with 10 lb traction. A standard posterior portal
was used with a 30° 5.0-mm arthroscope. Joint dis-
tension was accomplished with gravity flow from 3-L
bags. After 1995, joint distension was maintained with
an arthroscopic fluid pump maintained at 80 mm of
pressure. The drive-through sign was performed by
gently pushing the arthroscope through the joint be-
tween the humeral head and the glenoid at the level of
the anterior band of the inferior glenohumeral liga-
ment as described by Pagnani and Warren11,12 (Fig 1).
Care was taken not to scuff the articular cartilage of
the humeral head during this maneuver. Other intra-
operative findings included the presence of labrum
pathology, glenoid erosions, rotator cuff pathology,
and Hill-Sachs lesions. SLAP lesions were classified
using the classification of Snyder et al.19
The patients’ final diagnosis was determined based
on the preoperative and intraoperative findings. Those
patients with findings of instability received a stabili-
zation procedure, and this group was used as the
standard when the sensitivity, specificity, and accu-
racy were assessed. Included in the group of patients
with the diagnosis of instability were patients with
either anterior, posterior, or multidirectional instabil-
ity. The diagnosis of instability was based on a com-
bination of historical, examination, and arthroscopic
findings. Patients with a diagnosis of instability had to
have a history of a subluxation, dislocation, or internal
impingement type pain. On examination, subluxation
of the shoulder had to reproduce their symptoms of
pain or instability, or they had a positive anterior
apprehension test or relocation maneuver. At the time
of arthroscopy, they had to have findings that substan-
tiated the diagnosis of instability, such as a Hill-Sachs
40 E. G. MCFARLAND ET AL.

FIGURE 1. The drive-through

sign is performed by gently
pushing the arthroscope through
the joint between the humeral
head and the glenoid at the
level of the anterior band of the
inferior glenohumeral ligament.
Arthroscopic views from a pos-
terior portal showing the dis-
tance between the humeral head
and glenoid increasing as the
arthroscope passes through the
joint.

lesion, a Bankart lesion, glenoid erosions anterior or symptomatic problem for which they had a surgical
inferior, or findings of internal impingement. A ma- procedure was considered the primary diagnosis. For
jority of the patients with posterior instability had example, a patient with a rotator cuff tear and acro-
involuntary instability but could demonstrate the sub- mioclavicular arthritis who had a rotator cuff repair
luxations (i.e., had a voluntary component). The diag- and distal clavicle excision would be considered as a
nosis of instability was used as the dependent variable cuff tear patient in our analysis. In patients with in-
for statistical analysis. stability who had incidental cuff fraying and who had
Statistical analysis was performed using a standard a stabilization procedure, we considered instability as
statistics package (SPSS for Windows 1997, SPSS, the primary diagnosis.
Inc, Chicago, IL). The ␹-square test with Yates con- Of the patients, 234 (69%) had a positive drive-
tinuity correction and standard descriptive statistics through sign (Table 1). In the 76 patients with insta-
were performed. Logistic regression was used to ex- bility, 70 (92.1%) had a positive sign and 6 patients
amine the ability of the drive-through sign to predict (7.9%) a negative sign. For 263 patients with other
the diagnosis of instability. Significance was set at diagnoses, 164 (62.4%) had a positive sign and 99
P ⬍ .05. (37.6%) had a negative sign (Table 2). For the diag-
nosis of instability, the drive-through sign had a sen-
RESULTS
There were 186 male patients (53%) and 163 female TABLE 1. Prevalence of the Drive-Through Sign
patients (47%) with a average age of 43 years (range, in Study Patients
12 to 86 years). The final diagnosis was instability in
76, impingement or partial rotator cuff tear in 91, Diagnosis
full-thickness rotator cuff tear in 80, frozen shoulder Drive-Through Test Instability All Others Total
or adhesive capsulitis in 27, and “other” in 65 patients
(acromioclavicular pathology, synovial cysts, osteoar- Positive 70 164 234
thritis, rheumatoid arthritis, and avascular necrosis). Negative 6 99 105
Total 76 263 339
In patients with more than 1 diagnosis, the more
 ARTHROSCOPIC DRIVE-THROUGH SIGN 41

TABLE 2. Prevalence of Drive-Through Sign showed that the drive-through sign was not a good
by Diagnosis predictor of instability (odds ratio ⫽ .14).
Present Absent
DISCUSSION
Diagnosis n % n % N
This study shows that the drive-through sign is
Instability 70 92.1% 6 7.9% 76
(29.8%)* (5.8%)* associated with shoulder laxity but not with SLAP
Impingement and lesions. Although most shoulders with instability had
partial tear 62 68.1% 29 31.9% 91 a positive drive-through sign, the presence of a posi-
(26.5%)* (27.6%)* tive sign did not rule in that diagnosis. The drive-
Complete tear 44 55% 36 45% 80
through sign is very sensitive for the diagnosis of
(19%)* (34.6%)*
Frozen shoulder 11 40.7% 16 59.3% 27 instability and it rarely will be absent if instability is
(4.8%)* (13.5%)* present. However, the lack of specificity of the sign
Other 47 72.3% 18 27.7% 65 does not allow one to conclude that the presence of a
(18.7%)* (17.1%)* drive-through sign means the shoulder is unstable.
Total 234 69% 105 31% 339
Like many observations made at the time of arthro-
(100%)* (100%)*
scopy, the exact meaning of a positive drive-through
* Percent of column. sign currently must remain in question. There have
been no biomechanical studies that show what liga-
ments or structures must be damaged or cut to produce
sitivity of 92% and a specificity of 37.6%. The posi- a positive drive-through sign. There have been no
tive predictive value was 29.9% and its overall studies that explore the relationship of portal location,
accuracy was 49%. amount of arm traction, effect of fluid distension, or
There was association between younger age (12 to the amount of force required to produce a positive
40 years) and the presence of a positive drive-through drive-through sign. While the sign has been demon-
test (P ⬍ .001). There was no relationship between strated in patients undergoing arthroscopy in both a
this sign and the gender of the patients. There was an beach chair and a lateral decubitus position, the influ-
association between the drive-through sign and sev- ence of arm position on this test has not been eluci-
eral preoperative variables, including maximum active dated.11,12,13,15 This test may have some efficacy when
abduction of shoulder (P ⫽ .014), flexion of shoulder judging the result of an operative procedure on shoul-
(P ⫽ .027), apprehension sign (P ⫽ .005), and the der laxity, but this was not specifically studied in our
relocation test (P ⫽ .036). There was no association patient population.
between the drive-through sign and a positive preop- While this study shows that a positive drive-through
erative Neer or Hawkins impingement sign (Table 3). sign is not pathognomonic for shoulder instability, it is
For the intraoperative laxity examination, there was
a statistically significant relationship between a posi-
TABLE 3. Descriptive Analysis of Drive-Through Sign
tive drive-through test and higher degrees of shoulder With Other Study Variables
laxity (i.e., a Hawkins II or III) in either an anterior or
posterior direction. A significant association was also Variables ␹-Square DF P ⬍ .05
found between positive drive-through sign and the
Gender 0.295 1
degree of sulcus sign (i.e., grade II or III). There was Age 21.35 2 *
an association between a positive drive-through sign Apprehension 8.5 1 *
and glenoid erosions and Hill-Sachs lesions (P ⫽ Relocation 4.4 1 *
.018), but not with Bankart lesions (P ⫽ .071). There Instability 24.4 1 *
were 62 patients who had SLAP lesions (44 type I, 13 Bankart lesion 5.3 2
Hill-Sachs lesion 10.5 3 *
type II, 2 type III, and 3 type IV). There was no Rotator cuff pathology 16.6 4 *
association between SLAP lesions and a positive Anterior Hawkins sign 22.7 3 *
drive-through sign. There was a statistically signifi- Posterior Hawkins sign 21.8 3 *
cant relationship between the presence of a complete Sulcus test 25.06 3 *
rotator cuff tear and a positive drive-through sign Neer sign 2.88 1
Hawkins sign 1.25 1
(P ⫽ .002), but not between the sign and a partial tear
of the rotator cuff. The logistic regression analysis * P ⬍ .05. DF, degree of freedom.
42 E. G. MCFARLAND ET AL.
limited by the fact that only abnormal shoulders were
studied. The exact incidence and accuracy among
normal and abnormal shoulders would have to be
studied with cadavers or some other experimental
protocol. Also, it would be optimal for the test to be
repeated by other surgeons who may use slightly
different techniques when performing the drive-
through test.
In this study, we were unable to show a relation-
ship between SLAP lesions and a positive drive-
through sign. The incidence of SLAP lesions is
higher in this study than previous reports in the
literature, and this may be due to our patient pop-
ulation, which included degenerative lesions and
cuff tears. In our patients, type I lesions were the
most common whereas many studies report type II
as more common.15,19 Morgan et al.15 found a rela-
tionship between type II SLAP lesions, partial ro-
tator cuff tears, posterior-superior instability, and
anterior-inferior “pseudolaxity” (i.e., a positive
drive-through sign). In our study, there was no
relationship between the drive-through sign and
partial cuff tears, but there was a relationship to
full-thickness rotator cuff tears. This may have been
because of the small number of throwing athletes in
this study, in whom one might expect to see more
posterior-superior instability patterns associated
with partial rotator cuff tears.
Frozen shoulders that had been manipulated were
included in the analysis because we were initially
surprised to find that some of these patients had a
positive drive-through sign. It is possible that the
manipulation produced an abnormal laxity in their
shoulders. Statistical analysis of the data revealed
that the results and conclusions are not effected by
excluding the patients with frozen shoulder from
the analysis. If that group is excluded, the result are
essentially the same when studying the relationship
of the test to instability (sensitivity, 92%; specific-
ity, 35.7%; positive predictive value, 31.3%; nega-
tive predictive value, 93.4%; accuracy, 49%). Like-
wise, the relationship to laxity and SLAP lesions is
unaffected.
This study has particular importance when consid-
ering the factors that help make the diagnosis of
instability. In some patients, there are few signs of
glenohumeral instability. Although they may have
pain with their activities and sports, and while they
may have higher degrees of shoulder laxity, this study
show that a positive drive-through sign should not be
taken as the definitive or sole criterion for making a
diagnosis of occult instability. In patients undergoing
arthroscopy, while absence of the drive-through sign
makes the diagnosis of instability less likely, the pres-
ence of a drive-through sign is an unreliable indicator
for making the diagnosis of instability. Like many
findings at the time of arthroscopy, this sign should be
interpreted in the context of the patient’s history,
physical examination, and other arthroscopic findings.
The efficacy of this sign in evaluating shoulder laxity
before and after operative intervention warrants fur-
ther study.
Acknowledgment: The authors thank Mrs. Carie John-
son for invaluable assistance with this project.
REFERENCES
1. Answorth AA. Clinical evaluation of the unstable shoulder. In:
Warren RF, Craig EV, Altchek DW, eds. The unstable shoul-
der. Philadlephia: Lippincott-Raven, 1999.
2. Hawkins JH, Bokor DJ. Clinical evaluation of shoulder prob-
lems. In: Rockwood CA Jr, Matsen FA II, eds. The shoulder.
Philadelphia: WB Saunders, 1998;164-197.
3. Mallon WJ, Speer KP. Multidirectional instability: Current
Concepts. J Shoulder Elbow Surg 1995;4:54-64.
4. McFarland EG, Shaffer B, Glousman RE, Conway JE, Jobe
FW. Clinical and diagnostic evaluation of anterior shoulder
instability, impingement and rotator cuff tears. In: Jobe FW,
ed. Operative techniques in upper extremity sports injuries. ed.
St. Louis: Mosby, 1996;170-190.
5. Neer CS II. Dislocations. In: Shoulder reconstruction. Phila-
delphia: WB Saunders, 1990;273-341.
6. Neer CS II, Foster CR. Inferior capsular shift for involun-
tary inferior and multidirectional instability of the shoulder.
A Preliminary report. J Bone Joint Surg Am 1980;62:897-
908.
7. Silliman JF, Hawkins RJ. Classification and physical diagnosis
of instability of the shoulder. Clin Orthop 1993;291:7-19.
8. Cordasco FA, Pollock RG, Flatow E L, Bigliani LU. Manage-
ment of multidirectional instability. Oper Tech Sports Med
1993;1:293-300.
9. Foster R. Multidirectional instability of the shoulder in the
athlete. Clin Sports Med 1983;2:355-368.
10. Garth WP, Allman FL, Armstrong WS. Occult anterior sub-
luxations of the shoulder in non-contact sports. Am J Sports
Med 1987;15:579-585.
11. Pagnani MJ, Warren RF. Arthroscopic shoulder stabilization.
Oper Tech Sports Med 1993;1:276-284.
12. Pagnani MJ, Warren RF, Altchek DW, Wickiewicz TL,
Anderson AF. Arthroscopic shoulder stabilization using
transglenoid sutures. Am J Sports Med 1996;24:459-467.
13. Peterson CA, Altchek DW, Warren RF. Operative arthros-
copy. In: Rockwood CA, Matsen FA, eds. The shoulder. Ed 2.
Philadelphia: WB Saunders, 1998.
14. Jobe FW, Kvitne RS, Giangarra CE. Shoulder pain in the
overhead or throwing athlete: The relationship of anterior
instability and rotator cuff impingement. Orthop Rev 1989;18:
963-975.
15. Morgan CD, Burkhart SS, Palmeri M, Gillespie M. Type II
SLAP lesions: Three subtypes and their relationships to supe-
rior instability and rotator cuff tears. Arthroscopy 1998;14:
553-565.
16. McFarland EG, Campbell G, McDowell RN. Posterior shoul-
 ARTHROSCOPIC DRIVE-THROUGH SIGN 43

der laxity in asymptomatic athletes. Am J Sports Med 1996;
24:468-471.
17. Warner JJ, Micheli LJ, Arslanian LE, Kennedy R. Patterns of
flexibility, laxity and strength in normal shoulders and shoul-
ders with instability and impingement. Am J Sports Med 1990;
18:366-365.
18. Harryman DT II, Sidles JA, Harris SL, Matsen FA III.
Laxity of the normal glenohumeral joint: A quantitative
in vivo assessment. J Shoulder Elbow Surg 1992;1:66-
76.
19. Snyder SJ, Karzel RP, DelPizzo W, Ferkel RD, Friedman MJ.
Slap lesions of shoulder. Arthroscopy 1990;6:274-279.